Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    references to sacrificial practices or related concepts

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with sacrilege or sacrificial concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ITNESS
    -0.80
     Cheong
    -0.73
    è¦ļéĨĴ
    -0.71
    agher
    -0.70
     Turk
    -0.66
     Pebble
    -0.65
     Nanto
    -0.64
     Clarkson
    -0.63
    NING
    -0.63
    ITH
    -0.63
    POSITIVE LOGITS
    rament
    1.19
    ificial
    1.18
    rifice
    1.11
    char
    1.09
    het
    0.99
    iety
    0.96
    rum
    0.92
    ré
    0.91
    ros
    0.91
     sac
    0.90
    Act Density 0.008%

    No Known Activations